Noninvasive determination of respiratory ozone absorption: development of a fast-responding ozone analyzer

We developed a chemiluminescent ozone analyzer and constructed an ozone bolus generator with the eventual goal of using a bolus-response method to measure noninvasively the longitudinal distribution of ozone absorption in human lungs. Because the analyzer will be used to sample gases within a single breath, it must have a sufficiently rapid response to monitor changes in ozone concentration during a four-second breathing period, yet its sampling flow must be small enough that it does not interfere with quiet respiratory flows of 300 mL/sec. Our analyzer, which is based on the chemiluminescent reaction between 2-methyl-2-butene and ozone, has favorable performance characteristics: a 90 percent step-response time of 110 msec; a linear calibration from 0.03 to 10 parts per million (ppm)2 with a sensitivity of 2.3 nA/ppm; a signal-to-noise ratio of 30 evaluated at 0.5 ppm; and a minimum detection limit of 0.017 ppm. At an airflow corresponding to quiet breathing, the ozone generator is capable of producing single boluses with a peak ozone fraction as high as 4 ppm, but containing only 0.35 micrograms of ozone dispersed over a small volume of 19 mL. To test the combination of ozone analyzer and bolus generator, we performed bolus-response experiments at steady airflows of 50 to 200 mL/sec in excised pig and sheep tracheas. In spite of the small surface area available for radial diffusion, we found that 25 to 50 percent of the ozone introduced into the trachea was absorbed. By comparing the mathematical moments of the bolus input and the response curves to the predictions of a diffusion theory, we computed an absorption coefficient (K). The values of K increased with increasing airflow, implying that ozone absorption is limited by diffusion processes in the airway lumen as well as in the surrounding tissue.